Graham P. Wilson

Reconciling diverse lacustrine and terrestrial system response to penultimate deglacial warming in southern Europe

Unlike the most recent deglaciation, the regional expression of climate changes during the penultimate deglaciation remains understudied, even though it led into a period of excess warmth with estimates of global average temperature 1‒2 °C, and sea level ~6 m, above preindustrial values. We present the first complete high-resolution southern European diatom record capturing the penultimate glacial-interglacial transition, from Lake Ioannina (northwest Greece). It forms part of a suite of proxies selected to assess the character and phase relationships of terrestrial and aquatic ecosystem response to rapid climate warming, and to resolve apparent conflicts in proxy evidence for regional paleohydrology. The diatom data suggest a complex penultimate deglaciation driven primarily by multiple oscillations in lake level, and provide firm evidence for the regional influence of abrupt changes in North Atlantic conditions. There is diachroneity in lake and terrestrial ecosystem response to warming at the onset of the last interglacial, with an abrupt increase in lake level occurring ~2.7 k.y. prior to sustained forest expansion with peak precipitation. We identify the potentially important role of direct input of snow melt and glacial meltwater transfer to the subterranean karst system in response to warming, which would cause rising regional groundwater levels. This explanation, and the greater sensitivity of diatoms to subtle changes in temperature, reconciles the divergent lacustrine and terrestrial proxy evidence and highlights the sensitivity of lakes situated in mountainous karstic environments to past climate warming.

Explaining the past: abductive and Bayesian reasoning

Abductive reasoning is the central form of explanation in environmental reconstruction. This paper outlines the nature of the abductive method restating it in Bayesian terms, ie, in terms of a priori belief based on context and the relations between multiple proxies. A key role within this working method is played by evidence and, in particular, proxies of the past in the explanatory framework. Proxies, whether singularly or as groups, are interpreted within the context of the event, and the derived hypothesis, for which they have evidential claim. They are not considered as independent entities but rather as inter-related pieces of information concerning the likelihood of specific events. The degree of belief in an individual proxy is based on the interplay between the belief in the hypothesis before the proxy is considered and the relations and belief in the cluster of proxies of which the individual proxy is usually a part.